Tools like AlphaFold and ProteinMPNN have revolutionized how quickly sequences can be generated, yet the shift from digital design to physical testing remains a significant hurdle. Conventional, one-by-one experimental methods struggle to keep pace when candidate pools expand into the thousands. Tsingke’s new infrastructure addresses this by utilizing a staged process that prioritizes efficiency and scalability.
Tsingke Launches Workflow to Bridge AI Protein Design and Lab Testing
Beijing-based Tsingke has unveiled a high-throughput validation workflow designed to accelerate the transition from computational protein and antibody design to experimental verification. By integrating gene synthesis, parallel expression, and quantitative binding analysis, the system aims to resolve the bottleneck researchers face when testing thousands of AI-generated candidate molecules.
In the first stage, batch gene synthesis and parallel expression allow for the rapid screening of large candidate sets using ELISA-based binding assays. Once promising molecules are identified, the workflow moves to the second stage: scale-up production and purification. This allows for detailed characterization, including kinetic parameters like KD, kon, and koff, using technologies such as BLI or SPR. Tsingke’s platform currently supports processing up to 1,500 candidates daily, with gene-to-antibody delivery times as short as seven calendar days. By concentrating resources on top-performing molecules, the company intends to shorten the development cycle for therapeutic proteins and antibodies.
Comments (0)
No comments yet. Be the first!